Light-weight Calibrator: a Separable Component for Unsupervised Domain Adaptation

11/28/2019
by   Shaokai Ye, et al.
2

Existing domain adaptation methods aim at learning features that can be generalized among domains. These methods commonly require to update source classifier to adapt to the target domain and do not properly handle the trade off between the source domain and the target domain. In this work, instead of training a classifier to adapt to the target domain, we use a separable component called data calibrator to help the fixed source classifier recover discrimination power in the target domain, while preserving the source domain's performance. When the difference between two domains is small, the source classifier's representation is sufficient to perform well in the target domain and outperforms GAN-based methods in digits. Otherwise, the proposed method can leverage synthetic images generated by GANs to boost performance and achieve state-of-the-art performance in digits datasets and driving scene semantic segmentation. Our method empirically reveals that certain intriguing hints, which can be mitigated by adversarial attack to domain discriminators, are one of the sources for performance degradation under the domain shift. Code release is at https://github.com/yeshaokai/Calibrator-Domain-Adaptation.

READ FULL TEXT

page 2

page 7

page 8

page 11

research
07/14/2022

Prototypical Contrast Adaptation for Domain Adaptive Semantic Segmentation

Unsupervised Domain Adaptation (UDA) aims to adapt the model trained on ...
research
08/10/2023

SegDA: Maximum Separable Segment Mask with Pseudo Labels for Domain Adaptive Semantic Segmentation

Unsupervised Domain Adaptation (UDA) aims to solve the problem of label ...
research
10/17/2022

HyperDomainNet: Universal Domain Adaptation for Generative Adversarial Networks

Domain adaptation framework of GANs has achieved great progress in recen...
research
01/16/2019

A review of single-source unsupervised domain adaptation

Domain adaptation has become a prominent problem setting in machine lear...
research
07/12/2021

Source-Free Adaptation to Measurement Shift via Bottom-Up Feature Restoration

Source-free domain adaptation (SFDA) aims to adapt a model trained on la...
research
05/04/2023

ReMask: A Robust Information-Masking Approach for Domain Counterfactual Generation

Domain shift is a big challenge in NLP, thus, many approaches resort to ...
research
09/01/2023

Trust your Good Friends: Source-free Domain Adaptation by Reciprocal Neighborhood Clustering

Domain adaptation (DA) aims to alleviate the domain shift between source...

Please sign up or login with your details

Forgot password? Click here to reset